Unimodal Biometrics

In order to evaluate the performance of Markov chain based rank level fusion, experiments have been conducted. In the experiments, face, ear, and iris are used as the unimodal biometrics. Researchers investigated different biometric identifiers based on several factors including application scenario, associated cost and availability. The choice is highly personal and depends on the individual system requirements, resource availability, training schedule, and other factors.

Face appearance is a highly popular biometric because it has been a natural and a widely acceptable way for recognizing humans by other humans. Among all the biometric traits, face is the most common and heavily used biometric for person identification. Face recognition is friendly and non-invasive. The advantages of facial recognition include high public acceptance of this biometric, commonly available sensing devices, and the ease with which humans can verify the results.

Markov chain based rank level fusion method for multimodal biometric authentication system has been discussed. Definition of Markov chain and its construction mechanisms have been presented. Some early research on Markov chain has also been discussed. The main concentration was on two particular applications—for Web page ranking and for ranking similar items. Description of an extensive experimentation to evaluate the Markov chain method has also been presented. From the experiment, it can be observed that Markov chain approach is very promising and can outperform other fusion approaches in terms of biometric rank aggregation on virtual database.